Anal cancer is rare despite the fact that the anus joins the two most common sites of cancer in the human body, skin and colorectum. Anal cancer is more common in women than in men in all but one published series and in Kansas and western Missouri by a ratio of 3 to 1. Numerous predisposing factors have been suggested in published case reports including: condyloma accuminata, fistulae, lymphogranuloma inquinale, Crohn's disease extra-mammary Paget's disease, and various infections from anal intercourse among male homosexuals, to name a few. To data no epidemiological study of this rare cancer and etiologic factors has been published in the medical literature. Using the case-control methodology 100 women who have suffered anal cancer and 200 matched controls will be administered a tested interview instrument covering: previous disease of gut, skin, and female organs; cancer in patient and family; occupations and workplaces; obstetric and gynecologic history (including sexual practices); diet; and tobacco, alcohol, and drug use. In addition, a serum sample from each will be analized for herpes virus I and II antibody titers and fluorescent treponemal antibodies. Pathology material will be reviewed for diagnosis verification and to seek evidence of human papilloma virus infection Kansas and Missouri state venereal disease registries will be consulted for previous registration of cases and controls for any venereal disease. Cases will be obtained from records of the Cancer Data Service, the Kansas State cancer registry which also serves much of western Missouri. By using a method of Random Digit Dialing focused on cases' telephone exchanges it is hoped that a methodology for the case-control method of study of rare cancers can be developed, tested, and proved. Consultants from Biometry, Epidemiology, Pathology and Virology will participate directly in this three year study. In addition to performing the first epidemiologic study of anal cancer and exploiting the 3 to 1 ratio of females to males with this rare disease it is hoped that a methodology of control selection can be developed and provided so that this study will serve as a model for the study of etiology of other rare tumors with maximum efficiency and minimum expense.